Sampling via Stochastic Interpolants by Langevin-based Velocity and Initialization Estimation in Flow ODEs
This work addresses sampling challenges in machine learning and statistics, particularly for multimodal distributions, but it is incremental as it builds on existing stochastic interpolant and Langevin methods.
The paper tackles the problem of sampling from unnormalized Boltzmann densities by proposing a method based on probability-flow ODEs and linear stochastic interpolants, using Langevin samplers to generate intermediate samples and estimate velocity fields, with results showing efficiency on multimodal distributions and effectiveness in Bayesian inference tasks.
We propose a novel method for sampling from unnormalized Boltzmann densities based on a probability-flow ordinary differential equation (ODE) derived from linear stochastic interpolants. The key innovation of our approach is the use of a sequence of Langevin samplers to enable efficient simulation of the flow. Specifically, these Langevin samplers are employed (i) to generate samples from the interpolant distribution at intermediate times and (ii) to construct, starting from these intermediate times, a robust estimator of the velocity field governing the flow ODE. For both applications of the Langevin diffusions, we establish convergence guarantees. Extensive numerical experiments demonstrate the efficiency of the proposed method on challenging multimodal distributions across a range of dimensions, as well as its effectiveness in Bayesian inference tasks.